Browsing by Author "Ramanujam, Nirmala"

The American Cancer Society reported an estimated 300,000 new cases of breast cancer and 44,000 new breast-cancer related deaths in 2022 in the United States alone. With each new successfully treated primary tumor, there is a subsequent risk of disease recurrence. Recurrence poses a risk to 10% of patients within the first 5 years post treatment and a lifetime risk of 30% across all patients. While new tools are being developed to better understand and mitigate the risk of recurrence, triple negative breast cancers, which exhibit no targetable surface markers, offer little in the way of recurrence prediction or treatment. It is understood that tumor heterogeneity is a driving force in tumor recurrence. Temporal heterogeneity is associated with therapeutic treatment, where the administration either selects for resistance subpopulations of tumor cells that are able to recur or a de novo resistant phenotype arises that leads to recurrence. Additionally, it has been well documented that tumors vary spatially across a primary tumor. This heterogeneity takes the form of genetic, epigenetic, and phenotypic heterogeneity. One such phenotype of interest is metabolic heterogeneity. Metabolism is classified as a ‘Hallmark of Cancer’ and has been studied as a driver of tumor progression for almost a century since Otto Warburg first described the phenomenon of tumors exhibiting high rates of aerobic glycolysis. Optical imaging is well poised to study metabolic heterogeneity due to its ability to image cellular level features, to multiplex multiple endpoints, and the ability to image longitudinally. Endogenous fluorescence contrast of coenzymes NADH and FAD have been used to report on the redox state of in vivo tissue and distinguish cancerous from benign lesions. The Center for Global Women’s Health Technologies (GWHT) has employed the use of exogenous fluorescent contrast agents to provide substrate-specific metabolic information. Three fluorescent agents have been validated including: 2-[N-(7-nitrobenz-2-oxa-1, 3-diazol-4-yl) amino]-2-deoxy-D-glucose (2-NBDG), a glucose derivative that is able to report on glycolysis; Tetramethylrhodamine ethyl ester (TMRE), a cation that is selectively attracted to the charge gradient generated by the mitochondria during ATP synthesis, making it a reporter of OXPHOS; and Difluoro-5,7-Dimethyl-4-Bora-3a,4a-Diaza-s-Indacene-3-Hexadecanoic Acid (Bodipy FL C16), a long chain saturated fatty acid is taken up by the cell and undergoes beta oxidation similar to native fatty acids. More recently, GWHT has begun combing these fluorescence agents for in vivo use to provide a wholistic understanding of cancer metabolism. The work here sets out to develop a novel optical imaging platform that is capable of imaging multiplexed metabolic endpoints, for quantitative intra-image analysis of metabolic gradients. This technology is built on the use of exogenous fluorescence contrast agents to report on substrate or pathway specific axes of metabolism. By simultaneously introducing multiple contrast agents, it is possible to capture a more wholistic snapshot of tissue metabolism. To encourage the adoption of this technology, a novel low-cost instrument will also be developed. Leveraging a consumer grade CMOS camera and variable focus lens, it is possible to image over multiple length scales, capturing both bulk tumor features and also single cell features. The flexibility offered by this simple innovation will allow for metabolic imaging to be applied over a variety sample type. Three specific aims were proposed to realize this goal by developing methods of multi-parametric exogenous contrast and low-cost instrumentation for multi-scale imaging of tumor metabolic heterogeneity in preclinical models. Aim 1 validated and demonstrated a method for the simultaneous injection and measurement of Bodipy FL C16 and TMRE to report on lipid uptake and mitochondrial activity, two potentially interrelated axes of metabolism. To validate this method, three sets of experiments were performed to establish that the two probes do not exhibit chemical, optical, or biological crosstalk. Chemical compatibility was established using liquid chromatography. Briefly, high molar concentration solutions of each individual probe (Bodipy FL C16, TMRE, and 2-NBDG) were created alongside a solution of all three probes at the same concentration. Chromatograms were collected immediately upon mixing, after 1 hour and after 24 hours. The area under the curve for each probe at each time point displayed an area under the curve (AUC) within 2% of the AUC of the single probe solutions, suggesting no chemical reactions. Optical crosstalk was assessed using optical spectroscopy and tissue mimicking phantoms. Optical phantoms were created with tissue mimicking optical properties and various concentration of Bodipy FL C16, TMRE, polystyrene microspheres (tissue scattering mimic), and hemoglobin (tissue absorption mimic). Leveraging an inverse Monte Carlo algorithm, we demonstrated that accurate values for each fluorescent probe could be measured regardless of the concentration of the other optical probe or level of optical scattering or absorption, indicating optical compatibility. To address biological crosstalk, two sets of 4T1 tumor bearing mice were subject to optical spectroscopy with either 1) Bodipy FL C16 alone, 2) TMRE alone, 3) a dual injection of Bodipy FL C16 and TMRE. Fluorescence spectra were measured 2-, 4-, 6-, 8-, 10-, 20-, 30-, 40-, 50-, and 60-minutes post-injection to establish uptake kinetics. It was found that the uptake kinetics of the dual probe group were not statistically different from the single probe group, indicating biological compatibility. With no observable crosstalk between Bodipy FL C16 and TMRE, the two probes method was applied to characterize murine mammary gland and two tumor of differing metastatic potential (4T1 and 67NR). In addition, to Bodipy FL C16 and TMRE, oxygen saturation and total hemoglobin were extracted from estimates of optical absorption, and these 4 endpoints were used to attempt to cluster groups of tumor and normal tissue. Difficulty clustering tumor groups of varying metastatic potential suggest a need for imaging technology. In Aim 2 a low-cost fluorescence microscope was developed capable of performing quantitative fluorescence imaging over a variety of samples. The goal of this work was to design a system that could be adapted to image a number of different sample types include core-needle tissue biopsies, preclinical window chambers, and in vitro organoids. To accomplish this a low-cost CMOS detector was used with a variable magnification lens allowing for imaging at multiple length scales. Uniform illumination was a necessary criterion for quantitative imaging. To generate uniform illumination that could be scaled across multiple length scales, an LED coupled 1:4 fanout optical fiber was employed alongside a computational model to determine the positioning of each fiber. To automate the design of illumination, a computational model was employed where each optical fiber was modeled as a Lambertian emitter in a spherical coordinate system. To determine the ideal placement of each fiber such that the individual illumination contributions of all fibers summed to a uniform distribution, a global optimizer was employed. A genetic pattern search allowed for the selection of coordinates to produce uniform illumination that could be feasibly employed at the benchtop. This integrated system is referred to as the CapCell microscope. Using this computational approach, two uniform illumination profiles were designed, one with a high aspect ratio (length ≫ width) and one with a low aspect ratio (length = width). To demonstrate the utility of optimized illumination, core needle biopsies from 4T1 tumors were stained with a tumor-specific fluorescent contrast agent, HS-27 and imaged with either optimized or unoptimized gaussian illumination. The repeatability of intra-image features was compared for the two illumination scenarios, and it was found that uniform illumination repeatedly revealed the same fluorescent features across the sample. These features were further confirmed with standard histology. Window chamber imaging demonstrated the importance of designing application specific illumination. 4T1 mammary tumors were grown orthotopically before a window chamber was surgically implanted. Animals were injected with either Bodipy FL C16, 2-NBDG, or HS-27 and imaged with both the high AR and low AR illumination platforms. As expected, the low AR, designed for window chambers, had a higher power density at the sample site and thus increased contrast compared to the low AR images. With a method and a system in place, the goal of Aim 3 was to apply the optical imaging platform to observe spatiotemporal metabolic heterogeneity. To achieve this, the CapCell microscope was upgraded to enhance contrast and improve resolution for the visualization of capillaries and single cells. This was demonstrated using 4T1 window chamber models stained with acridine orange, a nucleus specific stain, and green light reflectance to highlight hemoglobin absorption in microvessels. Given the interplay between metabolism and vasculature it was desirable to employ a vessel segmentation approach to describe vascular features within an image. A Gabor filter and Djikstra segmentation approach was employed on metabolic images to enable metabolic and vascular comparisons across an image field of view. To test the improved CapCell system, 4T1 tumors were treated with combretastatin A-1, a vascular disrupting agent. Across the course of treatment, the CapCell was able to observe bulk changes in metabolism and vascular density. Additionally, by employing high resolution imaging, it was possible to observe relationships between each metabolic probe and vessel tortuosity. This analysis allowed for the identification of metabolically unique regions within each group of animals, demonstrating the ability of this technology to parse metabolically distinct regions of tumor. In total, the work outlined here describes the development of a novel optical imaging platform capable of quantifying intratumor metabolic heterogeneity of multiple metabolic endpoints over multiple length scales. The system expands on previous work developing methods for simultaneous measurement of exogenous fluorescent contrast agents to report on lipid uptake and mitochondrial activity. The system also introduced a novel computational approach to design uniform illumination for a low-cost microscope capable of imaging across multiple sample types. Together these technologies were used to observe metabolic heterogeneity in preclinical window chamber models following chemical perturbation. The technology introduced here, is primed for future exploration. First, it would be desirable to integrate all three exogenous contrast agents for simultaneous imaging of three axes of metabolism in vivo. Once accomplished, the sample technology could be applied to study metabolic and vascular changes associated with residual disease and tumors that are entering recurrence.

Background: India has the highest burden of cervical cancer mortality, globally, with 67,477 deaths in 2012. A novel device, the transvaginal digital colposcope (TVDC), or a small handheld colposcope, could potentially improve quality of care and address barriers to cervical cancer screening, by reducing patient discomfort and aiding practitioners in screening. Studies which validate India-WHO guidelines for cervical cancer screening report wide ranges of sensitivity and specificity for techniques currently used in low-resource settings, all of which are contingent on numerous factors from patient awareness to receptivity to user training, suggesting that the context is paramount to improving cervical cancer detection. To that end, assessment of the healthcare and physician environment in terms of practice and reaction to the new device is essential prior to device implementation in order to anticipate benefits or negative consequences of device use.

Methods: A survey was developed to explore experiences, practice, and approaches to cervical cancer screening based on a new technology, and administered to 15 gynecologists in various clinical settings in Hyderabad, India. First, participants answered questions about past and current practices for cervical cancer screening, diagnosis, and treatment procedures. Next, physicians assessed images from a clinical trial involving imaging of cervix by the TVDC and with standard colposcopy. To check physician interpretation of images from the clinical trial, biopsy or histologic confirmation was used for positive results, while colposcopy was used as the reference standard for negative results.

Results: Colposcopy and magnification for visualization of the cervix were preferred by all physicians, in spite of low frequency of in-house use or referrals for the procedure. Accuracy among physicians interpreting TVDC images ranged from 25%-100%, while accuracy with colposcopy images ranged from 38%-100%. Sensitivity for TVDC images and corresponding colposcopy images was 72% and 91% respectively, while specificity was 54% and 38% respectively. Physicians were more likely to report suspicion for cancer in positive cases with a false negative rate with TVDC images and corresponding colposcopy images at 19% and 0%. Images with the new device were either considered comparable to or were preferred to colposcopy images, but disagreement in interpretation between TVDC and colposcopy for the same patient ranged from 13%-63%.

Conclusion: This study shows how observation-based cervical cancer screening or diagnostic techniques, without preceding, adjunct screening or diagnostic tests, may have low specificity in disease detection. However, a new technology TVDC may be appropriate for this type of setting. Further research into patient attitudes, physician motivation, physician experience with colposcopy and clinical decision-making is required prior to implementation if gains in reduction of cervical cancer incidence and deaths are to be realized.

Cervical cancer is the second leading cause of death for women worldwide with 85% of deaths occurring in low and middle-income countries, despite being both preventable and treatable if detected early enough. The burden of disease persists primarily due to a lack of access to early diagnostics and significant proportion lost to follow-up. In Peru specifically, the mortality rates of cervical cancer are among the highest in the world with an annual incidence of 48.2 per 100,000.

To provide low-cost and accessible colposcopy while maintaining image quality, the point of care tampon like (Pocket) digital colposcope is being developed in the Tissue Optical Spectroscopy (TOpS) Lab at Duke University. As part of the ongoing Pocket colposcope development and validation, the Generation 3 Pocket colposcope was tested for concordance to the standard-of-care Goldway SLC-2000 digital colposcope in a pilot study conducted in Lima, Peru. The goal of this study is to demonstrate equivalence in clinical diagnostic performance of the Generation 3 Pocket colposcope versus the standard-of-care digital colposcope.

100 patients were enrolled under the IRB approved study protocol Pro00052865. Paired images of cervices were collected with the standard digital colposcope and the Pocket colposcope for each patient using acetic acid and Lugol’s iodine as contrast agents. Biopsies were taken as part of standard-of-care whenever required. The paired images were blinded by device, randomized, and sent with an electronic survey to three Duke affiliated physicians who are highly trained in colposcopy.

The primary outcome measured was level of agreement using an unweighted kappa statistic between 1) overall diagnosis and 2) Reid Colposcopic Index scores for the Generation 3 Pocket colposcope and Goldway colposcope. The secondary outcome measured was sensitivity, specificity, positive predictive value, and negative predictive value.

The percent agreement for all physicians combined between systems for the overall diagnosis was 83.78% with a kappa of 0.5786 and p-value of <0.0000, and the percent agreement for the Reid Colposcopic Index score comparison was 72.3% with a kappa of 0.4366 and p-value of <0.0000 indicating strong concordance. Both systems performed similarly when compared to gold-standard pathology, with level of agreement ~66% and a kappa statistic of ~0.3, and p-values <0.0000.

The Generation 3 Pocket colposcope performed similarly to the standard Goldway colposcope and can be used to increase access to colposcopy, thereby reducing the burden of cervical cancer morbidity and mortality in Peru and around the world.

Breast cancer outcomes globally are dependent on access to advanced operating room technology and radiation therapy facilities. In low-income countries, 90% of patients cannot access either radiation or surgery due to a lack of infrastructure, medical specialists, and funds. Therefore, there is a dire need for effective, resource-appropriate technology to improve cancer care in the absence of radiation and surgery, particularly for breast cancer which is the most common cancer in women globally.Breast cancer is a disease with a significant disease burden in both low- and high-resource settings. In both settings, breast cancer is fundamentally treated based on the degree of spread. Non-metastatic, focal tumors are treated with "local" therapy with or without additional "locoregional" therapy based on the degree of local invasiveness. When invasive tumors are found, treatment must include "systemic" immuno- or chemo-therapy as there is a presumed presence of circulating tumor cells. Some treatments, like radiation, occasionally incite an "abscopal effect" whereby tumor death in situ exposes tumor-associated antigens (TAAs), eliciting systemic, immune-mediated destruction of distant tumors; however, this mechanism remains elusive. An alternative "local" cancer therapy, ablation, involves focal destruction of tissue using a small instrument delivered under the skin with image guidance. Various ablation modalities (cryotherapy and radiofrequency ablation) have been observed producing the aforementioned abscopal effect because the necrotic/apoptotic tumor milieu remains in situ, activating tumor-specific cytotoxic T cells. Ablation with ethanol is particularly suited for low-resource settings as it can be performed with only a needle and syringe and may be guided with minimal imaging (ultrasound). Ablation with ethanol has been extensively used for hepatocellular carcinomas, and even though it is fast and effective, the injection of liquid ethanol into a dense tumor is difficult to control. Currently, ethanol ablation often requires multiple treatment sessions for residual or recurrent tumors. Here I utilized the phase-changing formulation of ethanol and the polymer ethylcellulose to increase coverage of a target ablation zone and produce a greater anti-tumor response. Previously, our lab has shown that ethyl cellulose-ethanol (ECE) ablation could more efficiently ablate superficial hamster cheek-pouch tumors than pure ethanol. However, a treatment strategy for breast cancer in low-HDI settings must address invasive disease which previous work with ECE had yet to address. In low-HDI settings, where there is less access to diagnostic services, many patients present with advanced disease. Even in high-HDI settings, current treatment options fall short for patients with recurrent or metastatic breast cancer. The goal of the dissertation was to develop a low-cost, easily accessible method for treating invasive breast cancers. To achieve this goal, I attempted to produce a reliable abscopal response using ECE ablation and other easily accessible drugs. I first optimized ECE ablation for use in a mouse breast cancer model, finding the maximum tolerable dose, and optimizing target tissue necrosis by modulating ethylcellulose concentration. I then characterized the local and systemic immune response to ECE ablation in several tumor models to identify a strategy for improving anti-tumor responses. To enhance the likelihood of an abscopal effect after ECE, I then utilized cyclophosphamide (CP) and buffer therapy to reverse tumor microenvironment (TME) hostility. Oral sodium bicarbonate buffer therapy (bicarb) reduces tumor acidosis and has been shown to increase cytotoxic T lymphocyte (CTL) infiltration into tumors and decrease CTL anergy. CP, a widely accessible chemotherapy, has immunomodulatory effects when used at low, non-curative doses, specifically depleting pro-tumor regulatory T cells. I demonstrated that an anti-tumor response after ECE ablation is more likely in a tumor primed with sodium bicarbonate and low-dose CP. I will refer to this combination treatment as ECE + CP + bicarb. To optimize the treatment and demonstrate efficacy, small animal tumor models were utilized to determine in vivo anti-cancer responses. Both non-metastatic and metastatic models were utilized to determine both the local and systemic response to the new therapeutic ECE + CP + bicarb to understand for which types of breast cancer this therapy was appropriate. Aim 1) Maximize breast tumor necrosis using ethanol ECE injections. First, I optimized ECE delivery to increase target tissue necrosis while minimizing adverse events and tumor growth. I used various dosing schedules to determine the maximum tolerable ECE dose in murine 67NR flank tumors, which is 6 mL/kg or 150 µL for a 25 g mouse. The concentration of ethylcellulose in ECE was modulated to determine the role of the phase-changing polymer on the target tissue ablation. I found that 6% ethylcellulose produced the most tumor necrosis and injectate retention at the injection site, thus 6% ECE was selected as the optimal concentration for these non-metastatic 67NR tumors. I also demonstrated that compared to ethanol alone, ECE improves the ablation zone's compactness and decreases local adverse events due to ethanol leakage. Using Raman spectroscopy through ex vivo tissue, I found that ECE slows ethanol diffusion through 67NR tumors compared to pure ethanol alone. Finally, I demonstrated that ECE improves long-term survival compared to an injection of the same volume of pure ethanol in murine tumors. While I developed a method of ECE that able to reduce primary tumor growth in a non-metastatic model, the local and systemic effect of ECE was still unknown. Aim 2) To characterize the local and distant immune response to ECE. To develop a therapy capable of treating invasive breast cancer, our goal was to create a systemic anti-tumor immune response initiated by tumor ablation. However, the immune response to ECE ablation had yet to be characterized. By comparing an injection of ECE to an injection of the same volume of saline in a mouse tumor model, the effect of ECE could be monitored. In this aim I demonstrated that ECE increases tumor-infiltrating lymphocytes in several models, including chemically-induced and cell-line derived tumors. Additionally, in mice lacking CD8+ T cells, the anti-tumor response of ECE was significantly reduced when compared to immunocompetent mice, suggesting reliance on CD8+ T cell immunity. In the metastatic 4T1 model, ECE increased splenic populations of activated CD8+ T cells and decreased the number of splenic CD11b+Ly6G+Ly6C+ neutrophils. Finally, I discovered that after a single ECE injection, the number of metastases were decreased compared to saline injections and standard of care treatment: surgical excision. Local ECE ablation was found to produce local and systemic immunomodulation favoring an anti-tumor immune phenotype; however, most primary tumors never completely regressed. Therefore, the readily-accessible, low-cost agents CP and bicarb were implemented to further enhance the anti-tumor immune response following ECE ablation. Aim 3) Enhancing ECE with readily-accessible, low-cost immunomodulatory agents. In Aim 2 the immune response to ECE ablation was characterized, however, it was not strong enough to cure animals with invasive TNBC. I hypothesized that ECE ablation was insufficient to cure malignant TNBC due to the highly immunosuppressive TME. Two methods for reducing TME immunosuppression were employed: low-dose CP and oral bicarb therapy. A single low-dose of CP was utilized to deplete Tregs before ablation. Bicarb was ingested by mice for the duration of the study to decrease tumor acidosis and increase the infiltration of anti-tumor T cells into the TME. TNBC cell lines with a range of natural immunogenicity were utilized to test the efficacy of ECE + CP + bicarb including 4T1, 67NR and EO771. The combination of ECE + CP + bicarb eradicated a majority of tumors, eliminating primary tumors and metastatic disease for most animals. Furthermore, the anti-tumor response was found to have a CD8+ T cell-dependent manner in EO771 tumors. In all three cell lines, mice cured with ECE + CP + Bicarb experienced a reduced tumor growth rate when re-challenged with a tumor. When surgery was used instead of ECE ablation, the antimetastatic effect was reduced implying that the in situ necrosis left by ECE ablation is crucial for the systemic anti-tumor response. In summary, I successfully created a novel anti-cancer therapeutic using ECE + CP + bicarb that is effective in aggressive TNBC tumors. The work in these Aims laid a foundation for the use of ECE ablation in breast tumors. A safe and effective dosing strategy was identified in small animal models, as well as methods for boosting the anti-tumor response to ECE ablation. An anti-tumor response to ECE ablation was identified along with the antimetastatic properties of local ECE ablation. These findings provoke many new research questions about the interplay of acidosis, wound healing, inflammation, and necrosis in the TIME and how they affect systemic disease progression. ECE ablation still requires much more investigation to reach the ultimate goal of impacting patient outcomes. For example, the mechanism for the anti-tumor and anti-metastatic response has yet to be fully elucidated. The work here suggests that CD8 T cells are implicated in the therapeutic response; however, the impact of ECE ablation on other crucial players in the TIME (myeloid cell populations, tumor metabolism, hypoxia, and the extracellular matrix) are largely unknown. Additionally, since the therapeutic power of ECE + CP + bicarb does not rely on specific tumor biomarkers, ECE + CP + bicarb could be effective in other tumor types. Specifically, we are interested in using ECE for cervical cancer which disproportionately affects low-HDI settings resulting in significant mortality globally. Another strategic use of the immunomodulatory effect of ECE is in combination with immunotherapies. ECE ablation induces a local inflammatory response and releases necrotic tumor debris that may increase the strength of the response to checkpoint inhibitors. Future research is needed to assess these new combinations.

Cervical cancer is the fourth most common female cancer in the world, primarily caused by the human papillomavirus (HPV). It about 570, 000 women annually, killing more than half of those affected. Approximately 90% of cervical cancer mortality occurs in low- and middle- income countries and this number is expected to rise to 98% by 2030, furthering global disparities in health care. Even within high-income countries where cervical cancer incidence and mortality has decreased drastically - by over 70% over the last 50 years - regional and racial disparities in cervical cancer mortality continue to exist among people with limited access to health care. The drastic decrease in incidence and mortality among high-income countries can be primarily attributed to a multi-tiered screening and treatment approach.

The American Society for Colposcopy and Cervical Pathology (ASCCP) recommends screening with the Papanicolau (Pap) smear/Cytology every 3 years from the age of 21-29 years. For women aged 30-65 years, HPV and Pap smear co-testing every 5 years, or HPV alone every 3 years is recommended. Positive results have a variety of follow up management recommendations. For a high-risk result, a Colposcopy-guided biopsy is performed to provide diagnostic conformation and classification of pre-cancer or cancer. Colposcopy involves application of contrast agents (3-5% acetic acid and/or Schillers iodine) onto the cervix. These sources of contrast produce changes in pre-cancerous lesions which are illuminated and magnified using a a colposcope. Biopsies (small pieces of tissue) are taken from the highlighted lesion regions and sent to a pathologist, which then informs treatment management.

This approach to screening, diagnosis and treatment, though highly effective, requires expensive equipment for cytology, colposcopy and histopathology analysis which may not be available in resource-limited settings. It also involves multiple visits for each tier with long wait times for results in between, leading to high rates of loss to follow-up among women.

Due to these limitations, the WHO recommends a screen-and-treat approach which involves a single visit with visual inspection of the cervix with the naked eye after application of acetic acid (VIA) followed by treatment with cryotherapy, which involves killing pre-cancerous cells with extremely low-temperatures or by LEEP for non-cryotherapy eligible women. Recently changes have been made to WHO guidelines recommending utilization of HPV screening in combination with or over the VIA test, where available followed by treatment. However due to limitations in cost and time to attain results associated with HPV tests, VIA is still recommended and used in most resource-limited settings.

VIA on its own is however not an appropriate measure due to lack of sensitivity, from visualizing cervix changes with the naked eye, lack of specificity associated with features which might also be seen in benign lesions and in other cervical trauma and inter-provider variability. Additionally, the lack of documentation with VIA limits quality control and opportunities for training. Finally, the speculum used not only in VIA but in all aspects of screening has been demonstrated as a barrier and associated with pain/discomfort, vulnerability and fear.

The goal of this work is to overcome these barriers using three main aims (1) develop a portable, low-cost cervix imaging device, the Callascope, that can be used without the speculum and potentially enable self-imaging for cervical cancer screening. (2) Validate the device through clinical studies and in-depth interviews, to determine feasibility of visualizing the cervix with a provider and by the patient, and (3) develop algorithms for automated classification of pre-cancer to reduce inaccuracies and inter-provider variability.

For aim 1, different designs were explored for an introducer which would replace the speculum, using 3D computer-aided design (CAD) software, and mechanical testing simulations were performed on each. Designs were rapidly prototyped and tested using a custom vaginal phantom across a range of vaginal pressures and uterine tilts to select an optimal design. Two final designs were tested on fifteen volunteers to assess cervix visualization, comfort and usability as compared to the speculum and the optimal design, a curved-tip inserter, was selected for subsequent use. Additionally, a smaller version of the Pocket Colposcope, a low-cost imaging device with image quality on par with higher end colposcope, was developed to accommodate the introducer and enable use without the speculum. Image quality assessment was performed to compare the smaller 2MP Pocket Colposcope, referred to as the Callascope to that of the original 5 MP Pocket Colposcope.

The final introducer device has a slim tubular body with a funnel-like curved tip measuring approximately 2.5-2.8 cm in diameter. The introducer has a channel through which the Callascope, a 2 megapixel (MP) mini camera with LED illumination fits to enable image capture. Mechanical finite element testing simulations with an applied pressure of 15 cm H2O indicated a high factor of safety (90.9) for the inserter. Testing of the device with a custom vaginal phantom, across a range of supine vaginal pressures and uterine tilts (retroverted, anteverted and sideverted), demonstrated image capture with a visual area comparable to that of the speculum for a normal/axial positioned uteri and significantly better than that of the speculum for anteverted and sideverted uteri (p<0.00001). A channel enabling application of liquid contrasts (Acetic acid or Lugol’s iodine) by spraying was developed and tested for ability to cover spray targets. Image quality testing with resolution and color targets demonstrated comparable image quality at low magnification.

For aim 2, 3 main clinical studies were conducted to assess the feasibility of the device as well as various in-depth interviews to understand attitudes towards the Callascope and acceptability for self-screening. For the clinical studies, a pilot was first conducted with fifteen volunteers through physician-assisted insertion to assess cervix visualization, comfort and usability as well as to optimize the design. Results showed adequate cervix visualization for 83% of patients. In addition, questionnaire responses from volunteers indicated a 92.3% overall preference for the inserter over the speculum and all indicated that the inserter was more comfortable than the speculum.

This was followed by a clinical study in which a physician imaged the cervix of patients using both the speculum and Callascope in a 2×2 crossover design for n=28 patients. The clinical study data indicated that the Callascope enabled similar visualization compared to the speculum while significantly improving patient experience. With physician insertion and manipulation, the Callascope enabled cervix visualization for 82% of the participants.

The third study involved a feasibility study with 12 volunteers for home-based self-cervix imaging with the Callascope. Eighty-three percent of participants were able to visualize their cervix with the Callascope on the first try and 100% after multiple attempts.

Finally, a clinical study on a small cohort of women was used to test the contrast application on the cervix. These studies indicated sufficient contrast application coverage over the entire cervix using 2ml of liquid in a 10ml syringe.

In-depth interviews were conducted in Durham to assess women’s knowledge about reproductive health, cervical cancer screening and attitudes towards the speculum and the Callascope. Themes from the interviews showed high lack of knowledge about cervical cancer screening, unfavorable attitudes towards the speculum examination procedure and overall preference for the Callascope over the speculum-based exam.

For the third aim, I developed a series of methods and algorithms for automated analysis of cervigrams (colposcopy images of the cervix) with various contrasts and a combination of the contrasts. First, I developed algorithms to pre-process pathology-labeled cervigrams and to extract simple but powerful color and textural-based features. The features were used to train a support vector machine model to classify cervigrams based on corresponding pathology for visual inspection with acetic acid (VIA), visual inspection with Lugol’s iodine (VILI), green illumination vascular imaging (GIVI) and various combination of these contrasts. The initial framework built on VIA and VILI achieved a sensitivity, specificity, and accuracy of 81.3%, 78.6%, and 80.0%, respectively when used to distinguish cervical intraepithelial neoplasia (CIN+) relative to normal and benign tissues. This is superior to the average values achieved by expert physicians on the same data set for discriminating normal/benign from CIN+ (sensitivity=77%, specificity=51%, accuracy=63%). For next steps, the methods used were extrapolated to a larger data set and results for VIA only, VIA+GIVI, VIA+VILI, and VIA+VILI+GIVI were explored. With additional contrast, diagnostic accuracy was increased. The results suggest that utilizing simple color- and textural-based features from VIA, VILI and GIVI images may provide unbiased automation of cervigrams, and that combining contrasts improved on only VIA use. This would enable automated expert-level diagnosis of cervical pre-cancer at the point-of-care.

Breast conserving surgery (BCS) is a common treatment option for breast cancer patients. The goal of BCS is to remove the entire tumor from the breast while preserving as much normal tissue as possible for a better cosmetic outcome after surgery. Specifically, the excised specimen must have at least 2 mm of normal tissue surrounding the diseased mass. Unfortunately, a staggering 20-70% of patients undergoing BCS require repeated surgeries due to the incomplete removal of the tumor diagnosed post-operatively. Due to these high re-excision rates as well as limited post-operative histopathological sampling of the tumor specimen, there is an unmet clinical need for margin assessment. Quantitative diffuse reflectance spectral imaging has previously been explored as a promising, method for providing real-time visual maps of tissue composition to help surgeons determine breast tumor margins to ensure the complete removal of the disease during breast conserving surgery. We have leveraged the underlying sources of contrast in breast tissue, specifically total hemoglobin content, beta-carotene content, and tissue scattering, and developed various fiber optics based spectral imaging systems for this clinical application. Combined with a fast inverse Monte Carlo model of reflectance, previous studies have shown that this technology may be able to decrease re-excision rates for BCS. However, these systems, which all consist of a broadband source, fiber optics probes, an imaging spectrograph and a CCD, have severe limitations in system footprint, tumor area coverage, and speed for acquisition and analysis. The fiber based spectral imaging systems are not scalable to smaller designs that cover a large surveillance area at a very fast speed, which ultimately makes them impractical for use in the clinical environment. The objective of this dissertation was to design, develop, test, and show clinical feasibility of a novel wide field spectral imaging system that utilizes the same scientific principles of previously developed fiber optics based imaging systems, but improves upon the technical issues, such as size, complexity, and speed,to meet the demands of the intra-operative setting.

First, our simple re-design of the system completely eliminated the need for an imaging spectrograph and CCD by replacing them with an array of custom annular photodiodes. The geometry of the photodiodes were designed with the goal of minimizing optical crosstalk, maximizing SNR, and achieving the appropriate tissue sensing depth of up to 2 mm for tumor margin assessment. Without the imaging spectrograph and CCD, the system requires discrete wavelengths of light to launch into the tissue sample. A wavelength selection method that combines an inverse Monte Carlo model and a genetic algorithm was developed in order to optimize the wavelength choices specifically for the underlying breast tissue optical contrast. The final system design consisted of a broadband source with an 8-slot filter wheel containing the optimized set of wavelength choices, an optical light guide and quartz light delivery tube to send the 8 wavelengths of light in free space through the back apertures of each annular photodiode in the imaging array, an 8-channel integrating transimpedance amplifier circuit with a switch box and data acquisition card to collect the reflectance signal, and a laptop computer that controls all the components and analyzes the data.

This newly designed wide field spectral imaging system was tested in tissue-mimicking liquid phantoms and achieved comparable performance to previous clinically-validated fiber optics based systems in its ability to extract optical properties with high accuracy. The system was also tested in various biological samples, including a murine tumor model, porcine tissue, and human breast tissue, for the direct comparison with its fiber optics based counterparts. The photodiode based imaging system achieved comparable or better SNR, comparable extractions of optical properties extractions for all tissue types, and feasible improvements in speed and coverage for future iterations. We show proof of concept in performing fast, wide field spectral imaging with a simple, inexpensive design. With a reduction in size, cost, number of wavelengths used, and overall complexity, the system described by this dissertation allows for a more seamless scaling to higher pixel number and density in future iterations of the technology, which will help make this a clinically translatable tool for breast tumor margin assessment.

Although two-thirds of the global cancer mortality burden is predicted to occur in low- and middle-income countries (LMICs), citizens of these countries have disproportionately less access to resources and facilities to provide effective care. Surgery, radiation therapy, and chemotherapy form the foundation of effective cancer care in high-income countries (HICs), but these modalities are largely unavailable in LMICs. Stemming from this disparity, long-term cancer survival rates are lower, and the mortality-to-incidence ratio is higher in LMICs. With limited healthcare spending and a large portion of expenditures out-of-pocket, non-communicable diseases such as cancer lead to financial catastrophe for millions of families annually and are a barrier to global development. To expand global access to cancer care and buttress the anti-cancer capabilities of overextended healthcare systems in LMICs, it is necessary to develop a therapy compatible with the constraints imposed by resource-limited settings.

To accomplish this goal, the work presented here describes a low-cost injectable ablative therapy suitable for widespread use in LMICs. This therapy is a modification of an existing technique entailing intratumoral injection of ethanol to induce necrosis of malignant cells (termed “ethanol ablation”) utilized to reduce tumor volume with either curative or palliative intent. Modifications are based on analysis of the mechanics of the injection process and entail the incorporation of the water-insoluble, ethanol-soluble polymer ethyl cellulose and reduction of the infusion rate and volume. Ethanol ablation is one of the original forms of tumor ablation, treatments in which the tumor microenvironment is altered via chemical or thermal means to destroy malignant tissue, and has achieved widespread clinical success in HICs. It is appealing for use in LMICs because it is low-cost, portable, electricity-independent, and minimally invasive. However, injected ethanol is highly pressurized and forms cracks within tissue leading to excessive leakage and an unpredictable distribution of injected ethanol, poor tumor coverage, and damage to adjacent organs. With the recognition of pressure-induced crack formation as a source of leakage, reducing the infusion rate and volume will improve localization. Further, the incorporation of ethyl cellulose is likely to reduce leakage because it forms a gel upon exposure to the aqueous tissue environment and reduces the permeability of fractured tissue. These innovations are poised to improve upon ethanol ablation while retaining its suitability for use in resource-limited settings.

Three specific aims were proposed to establish crack formation as a limiting factor for efficacy of ethanol ablation, characterize this novel tumor ablation technique and develop a framework for tailoring treatment protocols to specific lesion types and sizes. The first aim described the rheological properties of ethyl cellulose-ethanol and the gelling behavior upon exposure to water and found that reducing the infusion rate and incorporating ethyl cellulose decreased leakage in tissue-mimicking surrogates and improved ablative efficacy in chemically induced squamous cell carcinoma tumors in the hamster oral cavity. The viscosity of ethyl cellulose-ethanol solutions increases with the ethyl cellulose concentration, which has been found to improve localization of injected solutions. Further, as expected from a water-insoluble polymer, gel formation increases with higher ethyl cellulose concentrations and higher water-to-ethanol ratios as well. These findings motivate the use of higher ethyl cellulose concentrations and low infusion volumes, and indicate that gel forms upon injection as water diffuses into and ethanol diffuses away from the injection site.

Tissue-mimicking surrogates composed of agarose were utilized because they are transparent and poroelastic. This makes visualization of injected ethanol feasible in a material that replicates the dynamics of tissue’s mechanical response to infusion. In these surrogates, ethyl cellulose was demonstrated to reduce leakage and increase the distribution volume of injected ethanol, but only at moderate infusion rates. At infusion rates typically used in conventional ethanol ablation (approximately 100 mL/hr), excessive leakage was observed for pure ethanol and ethyl cellulose-ethanol alike. This result, taken in context with the established linear relationship between infusion pressure and rate, suggests that reducing the infusion rate is necessary to localize injected ethanol in addition to incorporating ethyl cellulose.

To demonstrate proof-of-concept of improved therapeutic efficacy, chemically induced oral squamous cell carcinoma tumors in the hamster oral cavity were utilized as they are similar to human primary tumors. Further, since they protrude from the surface of the oral cavity and injected fluid is not confined by adjacent tissue, they are susceptible to leakage and more difficult to treat. To evaluate conventional ethanol ablation in this model, high-rate (100 mL/hr) infusions were performed with an infusion volume 4x greater than the tumor volume. This protocol led to regression of only 4 of 13 treated tumors. However, with the reduction of the infusion rate to 10 mL/hr and infusion volume to a quarter of tumor volume, and the incorporation ethyl cellulose, 7 of 7 tumors regressed completely. In the absence of ethyl cellulose, reduction of infusion rate and volume led to regression of 0 of 5 tumors.

With the characterization of ethyl cellulose-ethanol and demonstration of proof-of-concept in Aim 1, the objective of Aim 2 was to investigate the role of infusion pressure in the mechanics of crack formation, as well as of ethyl cellulose in preventing leakage. Pressure-induced crack formation has been described to occur at a material-inherent critical pressure dictated by the fracture toughness and elasticity and can be quantified as the maximum pressure achieved during the infusion of air. In this aim, transparent tissue-mimicking surrogates were fabricated to match the critical pressure of ex vivo swine liver. To determine the relevance of the critical pressure, infusions were performed with two contrast agents dissolved in ethanol– one smaller than the surrogate pore size (fluorescein) and one larger (graphite). When the agarose pore structure was unfractured, only fluorescein was visible. After it was fractured, both contrast agents were visible. Using this system, fracture was observed to occur at the critical pressure and a modified technique to detect fractures via infusion pressure was established. While previous studies have demonstrated that fracture can be observed during the infusion, this is only possible with low-viscosity fluids unlike ethyl cellulose-ethanol. In these studies, it was demonstrated that unfractured agarose retains an elevated post-infusion pressure, but fractured agarose allows the pressure to dissipate rapidly. This result allows for non-invasive detection of crack formation in tissue during infusion of viscous fluids.

In ex vivo swine liver, as was the case in tissue-mimicking surrogates, crack formation was detected when the critical pressure was exceeded and increased leakage. In these studies, the injected ethanol distribution was determined by adding fluorescein to the injection solution, freezing tissue after the infusion, sectioning it, and imaging with a fluorescent microscope. Since the infusion pressure increases with rate and volume, this finding motivates the use of low rates and volumes when possible to improve localization. For low-volume infusions in which the pressure remained below the critical pressure, there was minimal leakage. While leakage, and the infusion pressure, increased with infusion rate (from 1 to 10 mL/hr) for pure ethanol, it did not increase for 6% ethyl cellulose-ethanol. The gel formation behavior of ethyl cellulose reduces leakage in the presence of infusion-induced cracks.

Having established proof-of-concept of ethyl cellulose-ethanol and its mechanism of action in localizing injected ethanol, the focus of Aim 3 was to characterize computed tomography (CT) imaging as rapid, non-destructive method to visualize injected ethanol, optimize the ethyl cellulose concentration, and investigate the relationship between the injected ethanol distribution and resultant extent of induced necrosis. Since ethanol is less attenuating of x-rays than water or tissue, it is readily visible with CT imaging. However, the accuracy of extraction of ethanol concentration from CT imaging has not yet been established. Utilizing ethanol-water mixtures as in vitro surrogates, the random and systematic components of measurement error were quantified, with the combined error defined as the root sum square of both components. The random error component arises from the variance of the radiodensity of a solution of fixed concentration. The systematic error component was quantified as the difference between the predicted and true radiodensity of ethanol-water mixtures, with the predicted value determined by a linear two-point calibration equation with pure water and ethanol at the extremes. The total measurement error was 13.4% with both components contributing approximately equal amounts. This error is low enough to confidently delineate between treated and untreated tissue.

Having established the utility of CT imaging to quantify the ethanol distribution volume, the ethyl cellulose concentration was optimized in ex vivo rat liver tissue submerged in buffer over a wider range of concentrations than has been feasible in previous models. The optimal ethyl cellulose concentration was defined as the formulation that maximized the volume of tissue infiltrated with a cytotoxic (> 20%) ethanol concentration. In these studies, 12% ethyl cellulose maximized the ethanol distribution volume by 8-fold in comparison to pure ethanol. It also led to the most spherical distributions as defined by the aspect ratio quantified as the ratio of the radius of gyration to the effective radius. These results were confirmed in in vivo rat liver in which 12% ethyl cellulose-ethanol yielded a distribution volume 3-times greater than pure ethanol.

In addition to improving localization of injected ethanol, 12% ethyl cellulose increased the extent of induced necrosis by 6-times in comparison to pure ethanol. Necrosis was quantified by excising treated tissue 24 hours post-ablation, cryopreserving, sectioning, and staining it with NADH-diaphorase. There was an approximate one-to-one equivalence of the ethanol distribution volume with the necrotic volume for 12% ethyl cellulose-ethanol. This validates the concentration-based thresholding strategy utilized to determine the ethanol distribution volume and confirms the utility of CT imaging. CT imaging is particularly appealing to assess the morphology of the ablative extent as three-dimensional reconstruction of the ablative extent from pathology is challenging. The equivalence between the distribution volume visualized with CT imaging and necrotic volume determined via pathology motivates further use of CT imaging in optimization of the ablation parameters. Pure ethanol had a necrotic volume of nearly half of the injected ethanol volume. While the comparison of this relationship between pure ethanol and 12% ethyl cellulose-ethanol was not statistically significant, it is indicative of prolonged exposure time achieved by ethyl cellulose that may be caused by delayed vascular clearance in vivo. This aim establishes CT imaging with concentration-based thresholding as a non-destructive, high-throughput method to optimize ablation parameters and tailor treatment to specific lesion types and sizes.

In conclusion, the objective of this work was to establish ethyl cellulose-ethanol ablation as an effective tumor ablation technique suitable for use in resource-limited settings with the goal of expanding global access to cancer treatment. In pursuit of this goal, aim 1 assessed the rheological and gelling behavior of ethyl cellulose-ethanol, established improved localization, and demonstrated proof-of-concept in treatment of chemically induced oral tumors. Aim 2 investigated the relationship between crack formation and infusion pressure, adapted an established model to detect crack formation by demonstrating that post-infusion pressure dissipation is characteristic of fractured tissue, and found that ethyl cellulose decreases leakage when cracks do form. Finally, aim 3 characterized the ethanol concentration measurement accuracy of CT imaging, optimized the ethyl cellulose concentration, and investigated the relationship between ethanol distribution volume and the resultant extent of induced necrosis. Ultimately, this work demonstrates that ethyl cellulose reduces leakage associated with ethanol ablation, improves therapeutic efficacy, and establishes a methodology for further optimization and to tailor treatment for specific applications.

The rate of occurrence and number of deaths associated with cancer continues to climb each year despite the continual efforts to battle the disease. When given a cancer diagnosis, it is particularly demoralizing and devastating news to a patient. Generally, cancer is defined as the uncontrolled rapid growth of abnormal cells with metastatic potential. In the cancer types originating from solid tissue or organ sites, a tumor will grow as a result of this rapid proliferation of cells. Surgical resection is a commonly used as part of the treatment regimen prescribed for these types of cancer.

Specifically in breast cancer, which impacts over 200,000 women a year, surgical intervention is used in almost 92% of treated cases. A specific surgical procedure is known as breast conserving surgery (BCS), where the physician removes only the tumor, while retaining as much normal tissue as possible. BCS is used in 59% of cases and is generally more preferable than the more radically mastectomy procedure where the entire breast is removed.

To minimize the chance of local recurrence, it is vital that the tumor is completely removed and residual cancer cells are not still present in the patient. This diagnosis is made by inspecting the edge of the resected tumor mass, typically known as the surgical margin. If tumor cells are still present at the margin, then a positive diagnosis is given and tumor cells likely remain inside the patient. Unfortunately, since margins are typically diagnosed using post-operative pathology a patient with a positive margin must undergo a second re-excision operation to remove additional tissue.

For breast cancer patients undergoing BCS, a staggering 20-70% of patients must undergo additional operations due to incomplete tumor removal during the first procedure.

Currently, there are two intra-operative techniques that are used, frozen section analysis and touch prep cytology. Although both have been proven to be effective in reducing re-excision rates, both techniques require

There remains a clinical unmet need for an intra-operative technology capable of quickly diagnosis tumor margins during the initial surgical operation

Optical technologies provide an attractive method of quickly and non-destructively assessing tissue. These techniques rely the interactions of light with tissue, which include absorption, scattering, and fluorescence. Utilizing proper measurement systems, these interactions can be measured and exploited to yield specific sources of contrast in tissue. In this dissertation, I have focused on developing two specific optical techniques for the purpose of surgical margin assessment.

The first is diffuse reflectance spectroscopy (DRS) which is a specific method to extract quantitative biological composition of tissues has been used to discern tissue types in both pre-clinical and clinical cancer studies. Typically, diffuse reflectance spectroscopy systems are designed for single-point measurements. Clinically, an imaging system would provide valuable spatial information on tissue composition. While it is feasible to build a multiplexed fiber-optic probe based spectral imaging system, these systems suffer from drawbacks with respect to cost and size. To address these I developed a compact and low cost system using a broadband light source with an 8-slot filter wheel for illumination and silicon photodiodes for detection. The spectral imaging system was tested on a set of tissue mimicking liquid phantoms which yielded an optical property extraction accuracy of 6.40 ± 7.78% for the absorption coefficient (µa) and 11.37 ± 19.62% for the wavelength-averaged reduced scattering coefficient (µs').

While DRS provided one potential approach to margin diagnosis, the technique was inherently limited in terms of lateral resolution. The second optical technique I chose to focus on was fluorescence microscopy, which had the ability to achieve lateral resolution on the order of microns. Cancer is associated with specific cellular morphological changes, such as increased nuclear size and crowding from rapidly proliferating cells. In situ tissue imaging using fluorescent stains may be useful for intraoperative detection of residual cancer in surgical tumor margins. I developed a widefield fluorescence structured illumination microscope (SIM) system with a single-shot FOV of 2.1×1.6 mm (3.4 mm2) and sub-cellular resolution (4.4 µm). The objectives of this work were to measure the relationship between illumination pattern frequency and optical sectioning strength and signal-to-noise ratio in turbid (i.e. thick) samples for selection of the optimum frequency, and to determine feasibility for detecting residual cancer on tumor resection margins, using a genetically engineered primary mouse model of sarcoma. The SIM system was tested in tissue mimicking solid phantoms with various scattering levels to determine impact of both turbidity and illumination frequency on two SIM metrics, optical section thickness and modulation depth. To demonstrate preclinical feasibility, ex vivo 50 µm frozen sections and fresh intact thick tissue samples excised from a primary mouse model of sarcoma were stained with acridine orange, which stains cell nuclei, skeletal muscle, and collagenous stroma. The cell nuclei were segmented using a high-pass filter algorithm, which allowed quantification of nuclear density. The results showed that the optimal illumination frequency was 31.7 µm−1 used in conjunction with a 4x 0.1 NA objective. This yielded an optical section thickness of 128 µm and an 8.9x contrast enhancement over uniform illumination. I successfully demonstrated the ability to resolve cell nuclei in situ achieved via SIM, which allowed segmentation of nuclei from heterogeneous tissues in the presence of considerable background fluorescence. Specifically, I demonstrated that optical sectioning of fresh intact thick tissues performed equivalently in regards to nuclear density quantification, to physical frozen sectioning and standard microscopy.

However the development of the SIM system was only the first step in showing potential application to surgical margin assessment. The nest study presented in this dissertation was to demonstrate clinical viability on a sample size of 23 animals. The biological samples used in this study were a genetically engineered mouse model of sarcoma, where a spontaneous solid tumor was grown in the hind leg. After the tumor was surgically removed from the animal and the relevant margin was stained with acridine orange (AO), a simple and widely available contrast agent that brightly stains cell nuclei and fibrous tissues. The margin was imaged with the SIM system with the primary goal of visualizing specific morphological changes in cell nuclei. To automatically segment AO-stained regions, an algorithm known as maximally stable extremal regions (MSER) was optimized and applied to the images.

As an intermediate step prior to diagnosing whole margins, a tissue-type classification model was developed to differentiate localized regions (75x75 µm) of tumor from skeletal muscle and adipose tissue based on the MSER nuclei segmentation output. A logistic regression model was used which yielded a final output in terms of probability (0-100%) the tumor within the localized region. The model performance was tested using an ROC curve analysis that revealed a 77% sensitivity and 81% specificity. For margin classification, the whole margin image was divided into localized regions and this tissue-type classification model was applied. In a subset of 6 margins (3 negative, 3 positive), it was shown that at a tumor probability threshold of 50% only 8% of all regions from a negative margins exceeded this threshold, while over 25% of all regions exceeded the threshold in the positive margins.

Variations in hemoglobin concentration can be indicative of a number of serious complications, including blood loss and anemia. Rapid, noninvasive measurements of hemoglobin are important in applications where blood status is reflective of patient well-being, such as in the emergency room, operating room, or the battlefield. Probe-based diffuse reflectance spectroscopy is capable of noninvasively quantifying tissue optical properties, including hemoglobin concentration. The quantification of hemoglobin concentration using optical methods is complicated by tissue scattering and the robustness of the algorithm and instrumentation used to interrogate the tissue. The sensing depth of diffuse reflectance spectroscopy can be tailored by the wavelengths of light and probe design used.

In this thesis, the accuracy and clinical viability of different diffuse reflectance spectroscopy implementations are presented. The robustness of an inverse Monte Carlo model, in which tissue optical properties are determined from measured reflectance using ultraviolet-visible (UV-VIS) wavelengths and a steady-state instrument, was tested using laboratory measurements. From the laboratory measurements, a set of references was identified which provided accurate absorption and scattering measurements, independent of the optical properties of the target. In addition, the ability to quantify hemoglobin concentration and saturation over large ranges and concentrations of multiple absorbers was established.

Following the laboratory measurements, a clinical study in which UV-VIS spectra were measured from the sublingual mucosa of patients undergoing surgeries was carried out. From this study, the correlations of extracted hemoglobin to expected blood hemoglobin were found to be improved when a simple ratiometric method based on isosbestic wavelengths of hemoglobin was used. During this study, the probe positioning in the mouth was found to be unwieldy, and so the transition to a more secure probe that could be taped to the hand was made.

In order to penetrate the overlying skin, near-infrared (NIR) wavelengths with a different probe geometry was explored. Further investigation of the inverse Monte Carlo model with NIR wavelengths was executed, and while in theory this combination should yield accurate optical property estimation, laboratory measurements indicated large errors, presumably due to the instrument or low magnitude and reduced spectral features of hemoglobin absorption in the NIR. Instead, the use of a well-established frequency-domain instrument coupled with diffusion approximation was implemented to measure spectra from the thenar eminence of volunteers undergoing induced hypovolemia and subsequent retransfusion. There were some moderate correlations with blood hemoglobin, but because both this method and the Monte Carlo method with mucosal probe placement showed higher variability with probe pressure than the isosbestic ratiometric method, further development of the ratiometric method was made.

The ratiometric method was developed using simulations and validated with phantoms and clinical data. Monte Carlo modeled reflectance was generated for a large range of biologically-relevant absorption and scattering values. The modeled reflectance was scaled by a calibration spectra obtained from a single laboratory phantom measurement so that linear regression equations relating hemoglobin concentration to ratios could be applied directly to clinical or laboratory measurements. Ratios which could best estimate hemoglobin concentration independent of saturation and scattering were determined through the simulation and laboratory measurements. Three isosbestic ratios - 545/390, 452/390, and 529/390 nm - were determined to best estimate hemoglobin concentration, and ratiometric-extracted hemoglobin was shown to correlate well to Monte Carlo-extracted hemoglobin in clinical measurements. Because only a single calibration measurement (which can be measured on a different day) is required per instrument and probe combination, this method can be implemented in near real-time and is thus appropriate for applications where hemoglobin concentration must be measured rapidly.

Cervical cancer disproportionately burdens women in low-resource settings, with over 85% of cervical cancer deaths occurring in developing countries due to lack of access to effective, high-quality screening programs that facilitate early detection and treatment. The aim of this study is to evaluate whether the performance of a transvaginal digital colposcope (TVDC) developed at Duke University is equivalent to the more expensive standard-of-care colposcope at identifying precancerous lesions of the cervix. Thirty-five paired cervix images, with confirmed pathologies and blinded by device, were sent electronically to six physicians, at four separate institutions, Duke University Medical Center (Durham, North Carolina, USA), La Liga Peruana de Lucha Contra el Cancer (Lima, Peru), Cancer Institute WIA (Chennai, India), and Kenyatta University (Nairobi, Kenya). Physicians completed a 1-page survey assessing cervix characteristics and overall severity of precancerous lesions for each image. Analysis included percent agreement between devices as well as identifying patterns across misdiagnosed images. The agreement between physicians using each device is 80.1% with kappa of 0.6049. The TVDC performed equivalent to standard-of-care colposcopy at identifying precancerous lesions of the cervix. Implications of these findings have the potential to create increased access to a culturally appropriate screening technology, thus reducing the burden of cervical cancer throughout the developing world.

Among women worldwide, cervical cancer is the third most common cancer with an incidence rate of 15.3 per 100,000 and a mortality rate of 7.8 per 100,000 women. This is largely attributed to the lack of infrastructure and resources in the developing countries to support the organized screening and diagnostic programs that are available to women in developed nations. Hence, there is a critical global need for a screening and diagnostic paradigm that is effective in low-resource settings. Various strategies are described to design an optical spectroscopic sensor capable of collecting reliable diffuse reflectance data to extract quantitative optical contrasts for cervical cancer screening and diagnosis.

A scalable Monte Carlo based optical toolbox can be used to extract absorption and scattering contrasts from diffuse reflectance acquired in the cervix in vivo. [Total Hb] was shown to increase significantly in high-grade cervical intraepithelial neoplasia (CIN 2+), clinically the most important tissue grade to identify, compared to normal and low-grade intraepithelial neoplasia (CIN 1). Scattering was not significantly decreased in CIN 2+ versus normal and CIN 1, but was significantly decreased in CIN relative to normal cervical tissues.

Immunohistochemistry via anti-CD34, which stains the endothelial cells that line blood vessels, was used to validate the observed absorption contrast. The concomitant increase in microvessel density and [total Hb] suggests that both are reactive to angiogenic forces from up-regulated expression of VEGF in CIN 2+. Masson's trichrome stain was used to assess collagen density changes associated with dysplastic transformation of the cervix, hypothesized as the dominant source of decreased scattering observed. Due to mismatch in optical and histological sampling, as well as the small sample size, collagen density and scattering did not change in a similar fashion with tissue grade. Dysplasia may also induce changes in cross-linking of collagen without altering the amount of collagen present. Further work would be required to elucidate the exact sources of scattering contrast observed.

Common confounding variables that limit the accuracy and clinical acceptability of optical spectroscopic systems are calibration requirements and variable probe-tissue contact pressures. Our results suggest that using a real-time self-calibration channel, as opposed to conventional post-experiment diffuse reflectance standard calibration measurements, significantly improved data integrity for the extraction of scattering contrast. Extracted [total Hb] and scattering were also significantly associated with applied contact probe pressure in colposcopically normal sites. Hence, future contact probe spectroscopy or imaging systems should incorporate a self-calibration channel and ensure spectral acquisition at a consistent contact pressure to collect reliable data with enhanced absorption and scattering contrasts.

Another method to enhance optical contrast is to selectively interrogate different depths in the dysplastic cervix. For instance, scattering has been shown to increase in the epithelium (increase in nuclear-to-cytoplasmic ratio) while decrease in the stroma (re-organization of the extra-cellular matrix and changes in of collagen fiber cross-links). A fiber-optic probe with 45° illumination and collection fibers with a separation distance of 330 μm was designed and constructed to selectively interrogate the cervical epithelium. Mean extraction errors from liquid phantoms with optical properties mimicking the cervical epithelium for μa and μs' were 11.3 % and 12.7 %, respectively. Diffuse reflectance spectra from 9 sites in four loop electrosurgical excision procedure (LEEP) patients were analyzed. Preliminary data demonstrate the utility of the oblique fiber geometry in extracting scattering contrast in the cervical epithelium. Further work is needed to study the systematic error in optical property extraction and to incorporate simultaneous extraction of epithelial and stromal contrasts using both flat and oblique illumination and collection fibers.

Various strategies, namely self-calibration, consistent contact pressure, and the incorporation of depth-selective sensing, have been proposed to improve the data integrity of an optical spectroscopic system for maximal contrast. In addition to addressing field operation requirements (such as power and operator training requirement), these improvements should enable the collection of reliable spectral data to aid in the adoption of optical smart sensors in the screening and diagnosis of cervical precancer, especially in a global health setting.

In 2011, an estimated 230,480 new cases of invasive breast cancer were diagnosed among women, as well as an estimated 57,650 additional cases of in situ breast cancer [1]. Breast conserving surgery (BCS) is a recommended surgical choice for women with early stage breast cancer (stages 0, I, II) and for those with Stage II-III disease who undergo successful neo-adjuvant treatment to reduce their tumor burden [2, 3]. Cancer within 2mm of a margin following BCS increases the risk of local recurrence and mortality [4-6]. Margin assessment presents an unmet clinical need. Breast tissue is markedly heterogeneous which makes identifying cancer foci within benign tissue challenging. Optical spectroscopy can provide surgeons with intra-operative diagnostic tools. Here, ex-vivo breast tissue is evaluated to determine which sources of optical contrast have the potential to detect malignancy at the margins in women of differing breast composition. Then, H&E images of ex-vivo breast tissue sites are quantified to further deconstruct the relationship between optical scattering and the underlying tissue morphology.

Diffuse reflectance spectra were measured from benign and malignant sites from the margins of lumpectomy specimens. Benign and malignant sites were compared and then stratified by tissue type and depth. The median and median absolute deviance (MAD) was calculated for each category. The frequencies of the benign tissue types were separated by menopausal status and compared to the corresponding optical properties.

H&E images were then taken of the malignant and benign sites and quantified to describe the % adipose, % collagen and % glands. Adipose sites, images at 10x, were predominantly fatty and quantified according to adipocyte morphology. H&E-stained adipose tissue sections were analyzed with an automated image processing algorithm to extract average cell area and cell density. Non-adipose sites were imaged with a 2.5x objective. Grids of 200µm boxes corresponding to the 3mm x 2mm area were overlaid on each non-adipose image. The non-adipose images were classified as the following: adipose and collagen (fibroadipose); collagen and glands (fibroglandular); adipose, collagen and glands (mixed); and malignant sites. Correlations between <&mus′> and % collagen in were determined in benign sites. Age, BMI, and MBD were then correlated to <&mus′> in the adipose and non-adipose sites. Variability in <&mus′> was determined to be related to collagen and not adipose content. In order to further investigate this relationship, the importance of age, BMI and MBD was analyzed after adjusting for the % collagen. Lastly, the relationship between % collagen and % glands was analyzed to determine the relative contributions of % collagen and % glands <&mus′>. Statistics were calculated using Wilcoxon rank-sum tests, Pearson correlation coefficients and linear fits in R.

The diagnostic ability of the optical parameters was linked to the distance of tumor from the margin as well as menopausal status. [THb] showed statistical differences from <&mus′> between malignant (<&mus′>: 8.96cm-1±2.24MAD, [THb]: 42.70&muM±29.31MAD) compared to benign sites (<&mus′>: 7.29cm-1±2.15MAD, [THb]: 32.09&muM±16.73MAD) (p<0.05). Fibroglandular (FG) sites exhibited increased <&mus′> while adipose sites showed increased [&beta-carotene] within benign tissues. Scattering differentiated between ductal carcinoma in situ (DCIS) (9.46cm-1±1.06MAD) and invasive ductal carcinoma (IDC) (8.00cm-1±1.81MAD), versus adipose sites (6.50cm-1±1.95MAD). [&beta-carotene] showed marginal differences between DCIS (19.00&muM±6.93MAD, and FG (15.30&muM±5.64MAD). [THb] exhibited statistical differences between positive sites (92.57&muM±18.46MAD) and FG (34.12&muM±22.77MAD), FA (28.63&muM±14.19MAD), and A (30.36&muM±14.86MAD). Due to decreased fibrous content and increased adipose content, benign sites in post-menopausal patients exhibited lower <&mus′>, but higher [&beta-carotene] than pre-menopausal patients.

Further deconstructing the relationship between optical scattering and tissue morphology resulted in a positive relationship between <&mus′> and % collagen (r=0.28, p=0.00034). Increased variability was observed in sites with a higher percentage of collagen. In adipose tissues MBD was negatively correlated with age (r=-0.19, p=0.006), BMI (r=-0.33, p=2.3e-6) and average cell area (r=-0.15, p=0.032) but positively related to the log of the average cell density (r=0.17, p=0.12). In addition, BMI was positively correlated to average cell area (r=0.31, p=1.2e-5) and negatively related to log of the cell density (r=-0.28, p=7.6e-5). In non-adipose sites, age was negatively correlated to <&mus′> in benign (r=-0.32, p=4.7e-5) and malignant (r=-0.32, p=1.4e-5) sites and this correlation varied significantly by the collagen level (r=-0.40 vs. -0.13). BMI was negatively correlated to <&mus′> in benign (r=-0.32, p=4e-5) and malignant (r=-0.31, p=2.8e-5) sites but this relationship did not vary by collagen level. MBD was positively correlated to <&mus′> in benign (r=0.22, p=0.01) and malignant (r=0.21, p=4.6e-3) sites. Optical scattering was shown to be tied to patient demographics. Lastly, the analysis of collagen vs. glands was narrowed to investigate sites with glands between 0-40% (the dynamic range of the data), the linear model reflected an equivalent relationship to scattering from % glands and the % collagen in benign sites (r=0.18 vs. r=0.17). In addition, the malignant sites showed a stronger positive relationship (r=0.64, p=0.005) to <&mus′> compared to the benign sites (r=0.52, p=0.03).

The data indicate that the ability of an optical parameter to differentiate benign from malignant breast tissues is dictated by patient demographics. Scattering differentiated between malignant and adipose sites and would be most effective in post-menopausal women. [&beta-carotene] or [THb] may be more applicable in pre-menopausal women to differentiate malignant from fibrous sites. Patient demographics are therefore an important component to incorporate into optical characterization of breast specimens. Through the subsequent stepwise analysis of tissue morphology, <&mus′> was positively correlated to collagen and negatively correlated to age and BMI. Increased variability of <&mus′> with collagen level was not dependent on the adipose contribution. A stronger correlation between age and <&mus′> was seen in high collagen sites compared to low collagen sites. Contributions from collagen and glands to <&mus′> were independent and equivalent in benign sites; glands showed a stronger correlation to <&mus′> in malignant sites than collagen. This information will help develop improved scattering models and additional technologies from separating fibroglandular sites from malignant sites and ultimately improve margin assessment.

According to the World Health Organization, there were over 2 million new breast cancer cases in 2018. This number is projected to steadily increase year after year. American Cancer Society projections for 2020 list the breast as the leading cancer site for new cancer cases in females, estimating breast cancer to represent 30% of all new cases and 15% of cancer-related deaths.

A leading cause of breast cancer deaths is due to tumor recurrence following therapy. These tumors can recur years, sometimes decades, after treatment from reservoirs of residual cells that persist in a dormant state. Conversely, the absence of residual invasive disease following adjuvant therapy constitutes pathological complete response (pCR) and is positively associated with long-term relapse-free survival. This risk for recurrence is higher for women with human epidermal growth factor receptor 2 (Her2+) breast cancer or triple-negative breast cancer (TNBC). Approximately 50-70% of Her2+ patients and 40-55% of TNBC patients who undergo standard therapy achieve pCR; however, in the remaining patients, only a partial response occurs, leaving residual disease and an increased risk of relapse.

To mitigate the cancer burden, years of research have focused on several common biological capabilities of cancer, deemed the Hallmarks of Cancer, including sustained proliferation, genome mutations, replicative immortality, resistance to cell death, and a deregulated metabolism. Several recent studies have further reported that this last hallmark, metabolism, may be vital to understanding the underlying behavior of dormant and recurrent tumors. Once understood, these changes in metabolic pathways, referred to as metabolic reprogramming, can be leveraged as vulnerabilities and allow for the development of strategies to eliminate residual disease or prevent residual tumor cells’ subsequent reactivation into full recurrence.

For nearly 100 years, increased aerobic glycolysis has been considered a feature of rapidly proliferating primary tumors. This occurrence, where cells continue to use the metabolic pathway where glucose is converted to lactic acid to release its stored energy and produce adenosine triphosphate (ATP) despite the presence of oxygen, has been termed the Warburg Effect. Because of this, physicians frequently use nuclear medicine directly imaging glucose uptake, fluorodeoxyglucose (FDG) Positron Emission Tomography (PET) imaging, for the diagnosis and staging of cancer. In addition to glycolysis, mitochondrial metabolism through oxidative phosphorylation has grown in recognition as an additional energy source for cancer cells. In mitochondrial metabolism, the tricarboxylic acid (TCA) cycle generates energy carriers to be used in the electron transport chain. Here, the mitochondrial membrane potential provides a gradient to produce large amounts of ATP. Additionally, the TCA cycle can rely on sources of carbon besides glucose alone. A steadily growing consensus points to other energetic sources, such as glutamine, amino acids, and lipids, that are key to survival, especially following environmental stress, treatment, or before migration and metastasis.

Though metabolic reprogramming underpins aspects of tumor dormancy and recurrence, currently, there are no techniques available to provide a systems-level approach to investigate the major axes of metabolism. Several techniques that offer insights into cellular metabolism exist, such as the Seahorse assay, metabolomics, and FDG-PET imaging. They, however, are limited to in vitro model systems, single-time point analyses of in vivo model systems, or single-endpoint analysis of in vivo model systems, respectively. Further, neither the Seahorse assay nor metabolomics can capture information about both the tumor and its native microenvironment. Therefore, there is an unmet need for a method to study metabolism at a spatial resolution that can elucidate the metabolic modulation of residual cell populations longitudinally and across in vitro and in vivo models.

Optical imaging is well-suited to address this gap in technologies owing to its ability to measure multiple metabolic endpoints non-destructively and repeatedly. The Center for Global Women’s Health Technologies has developed protocols for the use of two optical probes 2-[N-(7-nitrobenz-2-oxa-1, 3-diaxol-4-yl) amino]-2-deoxyglucose (2-NBDG) and tetramethylrhodamine, ethyl ester (TMRE), to image glucose uptake and mitochondrial membrane potential, respectively, in preclinical cancer models. These endpoints are superior to imaging of the endogenous fluorescence of NADH and FAD (referred to as the redox ratio) by providing a direct measure of a substrate (glucose uptake) and metabolic output (mitochondrial metabolism). This optical, metabolic imaging approach fills a critical gap that exists between in vitro studies on single cells (Seahorse Extracellular Flux Assay) and whole-body imaging (FDG-PET imaging) and is complementary to metabolomics and immunohistochemistry (IHC) with endpoints measuring the major axes of metabolism.

The work described here details an innovative platform to image changes in the metabolism of primary tumors, residual disease, and recurrent tumors using a Her2+ genetically engineered mouse model. This model exhibits key features of dormancy and mimics sustained use of targeted therapy to facilitate understanding of tumor biology and function, assess recurrence risk, and design therapies to mitigate residual disease and recurrence altogether. Imaging at a cellular level resolution will not only document acute metabolic changes following Her2 downregulation but also allow for metabolic imaging of dormant cell populations that are typically too small to study in human patients, typically referred to as no evidence of disease (NED) in humans. This platform will push metabolic studies of tumor dormancy further.

Three specific aims were proposed towards this ultimate goal to develop a multiparametric platform to characterize the metabolic reprogramming of preclinical cancer models.

Aim 1 establishes the functional flexibility of the fluorescent glucose analog 2-NBDG to measure glycolytic demand and the fluorescent cation TMRE to measure mitochondrial membrane potential to report on the metabolic changes that occur throughout tumor progression, dormancy, and recurrence. Using a genetically engineered mouse-derived three-dimensional in vitro mammosphere model allowed for metabolic endpoints to be captured across key time points. Doxycycline (dox) addition and withdrawal modulates expression of Her2, which is overexpressed in primary and re-activated mammospheres, and downregulated in regressing and dormant mammospheres. The mammospheres were characterized using immunofluorescence to confirm phenotype. Ki67 expression was high in primary and re-activated mammospheres, confirming a proliferative phenotype typical of both primary and recurrent disease presented in the clinic. On the other hand, short-term dox withdrawal resulted in increased cleaved caspase 3 (CC3) expression, confirming apoptosis due to Her2 downregulation. Finally, both Ki67 and CC3 expression were negative in dormant mammospheres, demonstrating a viable, but non-proliferative, steady-state phenotype.

Metabolic imaging revealed unique metabolic phenotypes across the tumor development stages that were consistent with the gold standard assays. While primary mammospheres, overexpressing Her2, maintained increased glucose uptake (“Warburg effect”), after Her2 downregulation, regressing and residual disease mammospheres appeared to switch to oxidative phosphorylation. Interestingly, in mammospheres where Her2 overexpression was turned back on to model recurrence, glucose uptake was lowest, indicating a potential change in substrate preference following the reactivation of Her2, re-eliciting growth. These findings highlight the importance of imaging metabolic adaptations to gain insight into residual and recurrent disease’s fundamental behaviors.

This work paved the way for similar studies in vivo using a mammary window chamber with the ultimate goal of informing the potential impact of metabolically-targeted therapies on tumor dormancy and recurrence.

In Aim 2, 2-NBDG and TMRE imaging was applied to in vivo mammary tumors as they transitioned from primary tumors, through regression and dormancy, to regrowth as recurrent tumors. Two tumor models varying in periods of dormancy (termed slow recurring and fast recurring tumors) were selected to characterize the importance of either axis of metabolism in the context of recurrent disease. When comparing the glucose demand and mitochondrial membrane potential levels between slow and fast recurring tumors, both sets of primary tumors behaved similarly to the primary mammosphere cultures: increased 2-NBDG indicating highly glycolytic tumors with low TMRE indicating little mitochondrial activity. Following acute Her2 downregulation, there was an increase of mitochondrial activity that remained relatively constant through regression, dormancy, and recurrence for both tumor types. However, glucose uptake varied between the two tumor types following Her2 downregulation. The mice bearing slow-recurring tumors showed a resurgence of glucose uptake during recurrence; conversely, the mice bearing fast-recurring tumors maintained decreased glucose levels continually following Her2 downregulation. Because the fast-recurring tumors did not have a meaningful change in glucose uptake during recurrence, it was hypothesized that the fast-recurring tumors might have reprogrammed to use fatty acids as a fuel source. Indeed, inhibiting fatty acid oxidation in these tumors resulted in increased glucose uptake during regression. Additionally, following this acute change in metabolism due to the inhibition of fatty acid oxidation, the tumor’s dormancy period prior to recurrence was prolonged, pointing to lipids as a crucial fuel source for residual disease and recurrence in aggressive breast cancer.

Aim 2 showed the importance of lipid metabolism in residual disease and recurrence. Additionally, other groups have also shown increased reliance on fatty acid oxidation in breast cancer residual disease following oncogene downregulation. Thus, Aim 3 established a method of visualizing long-chain fatty acid uptake in breast cancer murine models. Until now, the ability to monitor such uptake has been limited to in vitro and ex vivo approaches. Here, an imaging strategy that combines a fluorescently labeled palmitate molecule, Bodipy FL c16, and intravital, optical imaging was developed to measure exogenous fatty acid uptake. Because the palmitate’s 16th carbon is fluorescently labeled, immediate degradation of the Bodipy dye during fatty acid oxidation (β-oxidation) is prevented, allowing for fatty acid to be visualized through fluorescence imaging.

This technique was validated in two breast cancer models: a MYC-overexpressing transgenic triple-negative breast cancer (TNBC) model, previously reported to dramatically upregulate fatty acid oxidation intermediates, and the murine model of the 4T1 family, a group of sibling tumor lines with a reported wide range of metabolic phenotypes.

Using a genetically engineered mouse-derived xenograft allowed for fatty acid uptake levels to be captured during MYC-overexpression and following oncogene downregulation. Similar to the previously described genetically engineered model, this model used doxycycline addition and withdrawal to modulate MYC expression.

Through in vivo Bodipy FL c16 imaging, fatty acid uptake was found to be increased in MYC-high tumors. This model showcased two critically needed features for clinically relevant study of fatty acid uptake: 1) longitudinal metabolite tracking in a single animal shown through intra-animal decreases in fatty acid uptake following MYC-downregulation; and 2) providing a link between oncogene expression, which can be modulated therapeutically, and metabolic endpoints. This decreased uptake is indicative of a less aggressive state and correlates with a visible reduction in tumor volume. Additionally, this method found an increased fatty acid uptake in tumors with high metastatic potential, as well as the ability of the system to monitor inhibition efficacy, potentially allowing for therapeutic pharmaceutical testing of drug efficacy.

This fast and dynamic approach to image fatty acid uptake in vivo is a tool relevant to study tumor metabolic reprogramming or the effectiveness of drugs targeting lipid metabolism.

Targeting a tumor’s metabolic dependencies is a clinically actionable therapeutic approach, but identifying subtypes of tumors that are likely to respond remains difficult. The work presented here indicates that an optical platform to image 2-NBDG, TMRE, and Bodipy FL c16 longitudinally is well suited to characterize breast cancer residual disease and recurrence’s critical metabolic features and to pinpoint metabolic vulnerabilities for potential treatments. While the primary goal was to develop an imaging strategy for the unprecedented assessment of residual and recurrent disease at high resolution in in vitro and in vivo models, this innovation also fits within the broader framework of existing metabolic assessment techniques and provides a systematic way to connect in vitro studies to whole-body imaging within the context of preclinical pharmacology research.

Future work will focus on establishing a combined imaging strategy for simultaneous imaging of all three endpoints, transitioning imaging to a hand-held microscope for wide-spread adoption and rapid metabolic phenotyping of clinical samples, and integrating optical spectroscopy with this imaging platform to track the long-term effects therapy has on an individual tumor’s metabolism. The third will enable the ability to retrospectively look for changes in primary and regressing phenotypes that might foreshadow dormant behavior or the risk of early recurrence.

Cancer treatment regimens often include surgery, radiation, and chemotherapy. Though the World Health Organization (WHO) Essential Medicines List includes many globally accessible chemotherapies, surgery and radiation are inaccessible to 90% of patients in low- and middle-income countries (LMICs) due to lack of infrastructure, medical specialists, and funds. Novel treatment options, such as immune checkpoint inhibitors (ICIs), are increasing in use in high income countries (HICs), but can be prohibitively expensive for patients, especially in LMICs. Further, even when accessible in HICs, ICI therapies are not always effective. Breast cancers are especially non-responsive to ICIs. There is a compelling need to advance and/or enhance therapies in both HICs and LMICs. We have developed a novel ablation therapy that encases ethanol in a polymer local destruction of tumors. This proposal shows how we can adapt this for both scenarios as described in greater detail below.Ablation, the chemical or thermal destruction of tissue, is an alternative or adjunct to surgery and radiation because it is less expensive, less time intensive and minimally invasive. In HICs ablation is mainly used for local tumor control, but it can also induce immunomodulation that aids systemic response. When used in combination with chemotherapy or ICI therapy, it can target local and systemic responses. However, LMICs, which often lack access to surgery, also lack access to thermal ablation methods such as radiofrequency ablation (RFA), microwave ablation (MWA), and cryoablation due to cost, reliability of electricity. Further, they often lack trained physicians and personnel to maintain equipment. Even in HICs, thermal ablation is not always accessible or possible due to tumor location, exclusion criteria, or cost. Overall, to achieve clinical translation of this therapy it is essential to understand both: 1) the effect of delivery parameters on distribution and necrosis and 2) the potential for combination of novel ablative therapies with chemotherapy and ICI therapy to inform treatment and practice. Ethanol ablation is portable and low-cost, allowing it to overcome treatment barriers in LMICs. However, ethanol ablation has limited treatment efficacy due to poor ethanol localization and off-target leakage. Incorporating ethyl-cellulose (EC), an ethanol-soluble, water-insoluble polymer, to ethanol help mitigate these limitations. EC-ethanol (ECE) transitions from liquid to fibrous gel upon injection into tissue (in-situ gelation). This acts to sequester ethanol, reduces off-target leakage and, overall, can improve ablation efficacy. ECE has the potential to create a more predictable distribution of ablation, therefore I investigated the impact of key components affecting the delivery and therapeutic effect of ECE (Aim 1) and investigate the biological impact of ECE in combination with current clinical treatment paradigms and as a novel drug delivery agent (Aim 2). Pursuit of these aims was intended to elucidate the efficacy, safety, and predictability of ECE ablation for use in cancer treatment and inform eventual clinical translation of this technology. The outcome should demonstrate that ECE is safe for human use and exhibits pharmacological activity, bringing this technology steps closer to investigation in clinical trials. Research in this dissertation pursued a thorough understanding of key factors governing the therapeutic effects of ECE with goal of informing translation of ECE to a clinical setting. I assessed the effect of formulation and delivery parameters on the resultant distribution or leakage and on necrosis. This can lead to algorithms enabling clinicians to select optimal tools and delivery methods to maximize treatment efficacy. Further, adoption of ECE in the clinical setting cannot be achieved without a clear understanding of the healing response to ablation and the safety of the procedure. Thus, time course analysis of the wound healing response and treatment safety compared to traditional ethanol ablation is necessary. To assess these key components, we need a method for assessment that allows for real time visualization of the ablation. For optimization we can us a high resolution more expensive technology, such as computed tomography, with the intent of adapting methods for more accessible technologies like ultrasound in the future. I developed a method for utilizing CT and investigated delivery parameters in both small and large animal models. In addition to investigating larger scale models to inform clinical translation (Aim 1), I also assessed these key determinants of injections success in small animal models to inform the biological mechanisms of injection efficacy (Aim 2). This led to investigating the synergy of ECE with ICI therapy and modification of the ECE formulation as a cytotoxic drug carrier. In particular, ECE ablation has potential for synergy with combination therapeutics, specifically immunotherapies and chemotherapeutic agents. This could have high potential for impact in HICs where implementation of immunotherapies and intensive chemotherapy regimens is more common and accessible. ECE exposes tumor antigens to T cells, evoking an immune-stimulatory response. I hypothesized that ECE can prime “cold” tumors to enhance response to ICI, for which many breast cancers are non-responsive. Previous work demonstrated that low- dose cyclophosphamide enhances the therapeutic effect of ECE. Therefore the combination of ECE ablation and low-dose cyclophosphamide was a logical choice to investigate a neoadjuvant therapy to enhance response to ICIs in non-responsive tumors. I hypothesized that the in-situ gelation of EC can be implemented to improve intra-tumoral drug delivery. Ethanol is know for its cytotoxic effects on cells. vehicle compared to many inert polymer vehicles. Combining ECE with chemotherapy (often, small drug molecules) as a local treatment could synergize apoptotic and necrotic cell death induced by the drugs and ethanol, respectively, a therapeutic process absent in traditional drug carriers. Thus, I focused upon effect of ECE on small molecule transport, drug uptake and distribution throughout the body over time, and also assessed safety and efficacy of this novel combination treatment. The goal of this dissertation research was to improve efficacy, safety, and predictability of ECE ablation. I aimed to optimize delivery of ECE, working to understand the effects of salient injection parameters on distribution and necrosis. I also investigated ECE ablation in combination with chemotherapies or ICIs, helping to lay the groundwork for clinical translation, and informing the foundation for local and systemic treatment responses. To achieve this goal, I completed two parallel aims. Aim 1 focused on delivery of ECE, specifically development of real-time assessment methods, infusion parameter assessment at preclinical and clinical scales, and investigation of resultant necrosis and the wound healing response. Aim 2 focused on investigating the utility, efficacy, and safety of the ECE formulation as a cytotoxic drug carrier, and examined the synergy of ECE with ICIs.

Breast cancer is one of the leading causes of death every year in the United States for women. Breast conserving surgery (BCS) is one treatment option for these patients where achieving tumor-free surgical margins is desired to avoid local recurrence [1, 2]. Unfortunately, as many as 17.7-72% of patients undergoing BCS require repeat surgeries due to a close or positive surgical margin diagnosed post-operatively [3-11]. Histopathology is the current gold standard for determining surgical margin status; however, given the large volumes of resected breast tissue it is not feasible to section the entire specimen. High re-excision rates and limited histopathological sampling of the tissue represent a significant unmet clinical need for margin assessment for both the surgeon and pathologist. Quantitative diffuse reflectance (DR) spectral imaging has been shown to be a promising tool for interrogating tumor margins but patient and surgical factors have to be accounted for in order to fully exploit the discriminatory capability of the technology. The objective of this work was to characterize an instrument for margin assessment and to evaluate the effects of inter-patient variability and surgical and excisional factors on quantitative tissue optical properties, to devise strategies to exploit optical contrast for the detection of positive (<2mm) tumor margins. In addition, the performance of the spectral imaging platform was evaluated.

The DR spectral imaging device utilized in these studies consisted of a Xenon lamp, a multi-channel imaging fiber-optic probe, an imaging spectrograph, and a 2D charge-coupled device (CCD) [12]. The instrument was found to extract quantitative optical parameters related to tissue micro-morphology with <15% error. Cross-talk at the tissue surface was <1% when the spacing between adjacent channels was 10mm and the sensing depth of each channel was found to be 0.5-2.2mm, an appropriate depth for identifying close and positive tumor margins. Reproducibility of the imaging protocol was best when the probe was interfaced with lumpectomy specimens from the side; this methodology was maintained for all measurements from lumpectomies in this dissertation.

DR spectral images were acquired from lumpectomy margins and converted into composition maps of quantitative optical parameters. Mammographic breast density was found to have the greatest impact on the optical data with β-carotene concentration ([β-carotene]) and the ratio of [β-carotene] to the wavelength-averaged reduced scattering coefficient from 450-600nm (<μs'>) being significantly higher in the negative margins of high-density patients (p=0.017 and p=0.038, respectively). We originally hypothesized that increased [β-carotene] would be associated with an increase in fat; however the significant increase in [β-carotene] cannot be attributed to differences in the percentage of adipose tissue since low-density patients should theoretically have higher percentages of this tissue type. Hematoxylin and eosin analysis of the adipose sites (n=25) showed increased [β-carotene] (p=0.066), increased adipocyte density (p=0.034), and smaller adipocyte sizes (p=0.051) in the adipose tissues (where β-carotene is stored) of high-density patients. This analysis suggests that increased [β-carotene] is associated with smaller adipocytes and that high-density breasts overall have smaller adipocytes, thus affecting optical contrast. This increase in [β-carotene] actually served to increase contrast between negative and positive margins which resulted in better classification accuracy in the high-density patients with a conditional inference tree model (77% in low-density and 80% in high-density).

If the purpose of the spectroscopy tool is to provide a differential diagnosis of benign versus malignant tissue, there must be an understanding of how excision of the tissue affects the optical properties over time, and how differences in surgical techniques affect optical properties. DR spectra were acquired 17±4 minutes post-excision from 12 incised mastectomies and from the surface of 10 lumpectomies 7±3 minutes post-excision. A linear longitudinal model was used to fit the data and obtain a rate of change for the tissue parameters. In lumpectomies, [β-carotene], <μs'>, and [β-carotene]/<μs'> had the lowest percent change (<14%) over 30 minutes; total hemoglobin concentration ([THb]) and [THb]/<μs'> had higher percent changes (>40%) over 30 minutes; hemoglobin saturation (HbSat) showed non-linear changes making it a poor variable for ex vivo margin assessment; and Lymphazurin^TM concentration (denoted as [Lymphazurin^TM]) changed more than 200% in 30 minutes. Although the percent error in [Lymphazurin^TM] was high, all other tissue parameters could be quantified with <3.3% error even when Lymphazurin^TM was 80μM. No significant difference was found between benign and malignant rates of change, and baseline values were not significantly correlated with elapsed time post-excision. Initial values from benign non-cauterized mastectomy (n=13) and cauterized lumpectomy (n=59) sites were compared to assess the effect of cautery. [THb] was the only parameter that was significantly higher in the cauterized lumpectomies (p=0.013) compared to non-cauterized mastectomies.

The work in this dissertation shows the feasibility of using an optical device for margin assessment and that [β-carotene] and [β-carotene]/<μs'> emerge as important variables for differentiating negative and close/positive margins. These two parameters were likely most important since they were least affected by kinetics, cautery, and the presence of Lymphazurin^TM.

Breast cancer is the most diagnosed cancer and the second leading cause of mortality in women worldwide. With early access to screening and diagnosis, breast cancer patients in high-income countries (HICs) have a more than 90% five-year survival rate. Unfortunately, the breast cancer mortality rate is much higher in low- and middle-income countries (LMICs), even though the breast cancer incidence rate in these regions is lower than HICs. In LMICs, most women with breast cancer present with late-stage disease, which is associated with poor prognosis and a high mortality rate. One reason that has been attributed to this situation is delayed diagnosis due to a lack of medical resources such as centralized laboratories and access to pathologists. Gold standard breast cancer diagnosis in HICs relies on histological analysis of CNBs. Unfortunately, this procedure requires labor-intensive sample preparation and time-consuming evaluation, resulting in long turn-around time and extensive infrastructure. A rapid, low-cost diagnostic tool is needed for breast cancer patients in LMICs. However, in HICs, breast cancer also poses a burden in breast cancer treatment. Patients undergo breast-conserving surgery to minimize treatments but often suffer from secondary excision due to a lack of standardized intraoperative margin assessment methods. Currently, histological analysis on the excised tumor is the only post-operative margin assessment to determine the completeness of BCS. With centralized laboratories and access to pathologists, post-operative margin assessment in HICs still requires a few days to complete. Therefore, under current practice, breast cancer patients with a positive tumor margin from post-operative margin assessment will require additional surgery to minimize the risk of local reoccurrence. Intraoperative margin assessment is essential to reduce re-excision rates by providing an on-site evaluation of tumor margins and giving real-time feedback to clinicians to guide additional shavings in the same surgery. Therefore, there is an unmet need for a standardized, automated procedure that analyzes patients’ ex vivo tissue specimens similarly to traditional histology so that it provides rapid, low-cost diagnosis for patients in LMICs and serves as intraoperative margin assessment in HICs. Our lab previously developed a fast, low-cost molecular diagnostic platform to discriminate between malignant and benign breast CNBs. This strategy leverages the specific expression of heat shock protein 90 (Hsp90), a chaperone protein that is overexpressed on the surface of breast cancer cells. We established a non-invasive and rapid molecular imaging approach to quantify Hsp90 expression by using the FITC-tethered Hsp90 inhibitor (HS-27) as a fluorescent probe that binds surface Hsp90 receptors of breast cancer cells. Our preliminary clinical study showed that Hsp90 surface expression, as detected by HS-27, differentiates malignant tumors from normal breast tissue obtained by CNB. However, our study also showed that the contrast between cancer and benign tumor spanned a wide range. To further optimize contrast, it is important to identify the potential sources of systematic error that confound contrast and develop strategies to mitigate or improve upon them. To achieve this goal, I first studied both sources of Hsp90 contrast and potential confounders. Even though other groups have demonstrated in vitro, in vivo, and ex vivo imaging of HS-27, previous work has yet to be done to understand how viability and probe diffusion impact contrast. Here, I investigated how three sources of systematic error affect ex vivo Hsp90 imaging in preclinical models. Specifically, I explored the effect of tissue viability, molecular probe diffusion kinetics, and contact vs. non-contact Hsp90 imaging methods as potential error sources. I used 4T1 mammary tumors grown in a nude mouse model, from which core needle biopsies are obtained to emulate the clinical scenario yet provide a controlled environment to test these variables. My results were critical in identifying the window in which variables confound Hsp90 contrast. These studies also provided a methodology for optimizing point-of-care molecular imaging of tissue biopsies with other agents. The contact method that was used for Hsp90 imaging had a small field of view and thus multiple placements must be made to cover the entire core needle biopsy. Repeated removal and placement of the probe across the sample and the variable pressure associated with the placement was significant source of error. Therefore, I developed a wide-field, high-resolution portable microscope that can perform two-contrast simultaneous brightfield and fluorescence imaging. I used a fiber optic-based approach that allows for multiple points of illumination. A computational model was developed to simulate the illumination distribution on the sample plane. An optimizer was used to find the best spatial position of optical fibers with different field-of-view inputs. The novel computational approach allowed for the design of sample-specific uniform illumination across various samples and fields of view in a modular manner. I demonstrated the attributes of this approach through the design of a single device, the CapCell, that can be used for imaging a wide range of samples, including breast core needle biopsies (high aspect ratio sample), mammary tumor window chamber models, and tumor organoids (low aspect ratio samples). Importantly, the new technology could be used to image the entire sample in a single shot. Preclinical biopsies stained with HS-27 were used to validate the feasibility of Hsp90 imaging with this system. Uniform illumination increased the consistency of analyzing regions of interest within an image, enabling intra-image analysis and interpretation of local features that correspond with the presence of tumor cells in the corresponding histology image. The platform enabled a single system to image across anatomical locations and sample types of different sizes and geometries. The computational-based illumination model allowed for the design of uniform illumination systems that are independent of the detector. Changing the field of view being optimized ensured the design of custom systems so that the system can be adapted to image varying-sized biological specimens at fields of view and different spatial resolutions. Research has shown that brightfield images of breast tissue specimens could provide additional contrast to segment the specimen into its main components (adipose, collagen, and epithelium) using intrinsic optical properties of the tissue. Almost all breast cancer originates from epithelial cells, while adipose and collagen tissue together can make up more than half of breast tissue specimens. Thus, it is unnecessary for breast cancer diagnosis to have a detailed evaluation of adipose or collagen tissue. We hypothesized that excluding adipose and collagen tissue from ex vivo tissue analysis can improve breast cancer pathological analysis's time efficiency and accuracy. To this end, I developed a machine learning algorithm for automated breast tissue segmentation on breast CNB images. I utilized a multi-contrast brightfield imaging system to acquire brightfield CNB images using white-light and green-light illuminations. I performed color space transformation of acquired RGB images to extract color channels that were relevant to different tissue types. Extracted color channels were combined through multiplications and additions to generated colormaps that correspond to the different tissues. K-means clustering taking in ratios of white-light colormaps to green-light colormaps was used for tissue segmentation. I developed a two-step segmentation method by first segmenting adipose regions from original CNB images and then performing the tumor segmentation. I systematically analyzed the adipose and tumor segmentation performance using combinations of colormaps generated from extracted color channels. With the need to provide an alternative to traditional histology used for CNBs diagnosis and margin assessment, I aimed to leverage surface Hsp90 expression to develop a rapid, low-cost diagnostic tool to inform breast cancer on ex vivo specimens. The work starts with a systematic analysis of sources of errors in Hsp90 molecular imaging on ex vivo tissue specimens. A novel optical imaging technique is developed to accommodate the need for high-resolution, wide-field imaging of CNBs in order to minimize the effect of probe contact and pressure. Machine learning techniques have been explored for automated breast tumor segmentation on clinical CNBs. All these advances can be integrated into a portable, automated technology that can leverage bright field and fluorescence (Hsp90) images for rapid segmentation and classification of tumors in the context of diagnostic biopsy and/or margin assessment.

With the widespread adoption of mammograms for early breast cancer detection in high income countries (HICs), modern research has principally pivoted towards a focus on reducing overtreatment of patients, particularly those with early stage breast cancer. There are numerous examples of clinical practices and regulations that reflect this shift, including changes in screening recommendations, patient monitoring, re-excision guidelines, and genetic testing, all of which seek to reduce unnecessary intervention without compromising patient outcomes.

Despite changing guidelines, there remains a distinct lack of technologies to reduce overtreatment while ensuring the best possible outcome for patients. One such example is Breast Conserving Surgery (BCS) followed by radiation therapy. There is a wide-range of re-excision rates reported in the literature, but most groups report that 20-40% of patients undergo at least one re-excision. Taking additional shavings during BCS, new guidelines dictating relationships between margin status after BCS and re-excision, and radiation therapy all strive to maximize removal of residual tumor cells with as few surgeries as possible in patients with a new breast cancer diagnosis. However, secondary cancers from radiation therapy, the potential for cancer dissemination as a result of re-excision surgeries, and the burgeoning costs of repeat visits and interventions to an already depleted health care system necessitate new and innovative solutions to improve health outcomes while reducing health expenditures.

While seeking to improve patient experience in HICs, many women in low to middle income countries (LMICs) are unable to access adequate screening and life-saving treatments. Even those who manage to receive a proper diagnostic test are all too frequently lost to follow up care, leading to a disproportionately high burden of breast cancer mortality in LMICs.

The goal of the work presented here is to reduce overtreatment in HICs while simultaneously minimizing access barriers to screening and treatment for women in LMICs through developing a rapid and low-cost molecular diagnostic platform for breast cancer. In HICs, this diagnostic platform could be deployed at two points in the breast cancer care cascade: 1) during diagnostic biopsy to ensure adequate lesion sampling at the point-of-diagnosis, and 2) during intraoperative margin assessment as a ‘see-and-treat’ paradigm that utilizes a single agent to guide surgical resection and to treat residual disease during surgery. In LMICs where limited access to tissue processing equipment and a pathologist often render histological examination of tissue impossible, the diagnostic platform could be used to cheaply and robustly diagnose tissue at the point-of-care.

Three specific aims were proposed to develop the diagnostic platform. The first aim was to demonstrate a single-agent see-and-treat paradigm in pre-clinical models of breast cancer using a fluorescent tracer across all subtypes of breast cancer. The diagnostic piece began by showing that a fluorescently-tethered Hsp90 inhibitor (HS-27), made up of an Hsp90 inhibitor previously used in clinical trials tethered to a fluorescein isothiocyanate (FITC) derivative, is taken up by breast cancer cells in vitro regardless of receptor subtype, and that blocking the ATP binding pocket of Hsp90 leads to reduced HS-27 fluorescence, confirming that fluorescence is a result of HS-27 bound to its target. The in vitro study was expanded to define the therapeutic potential of HS-27 by demonstrating degradation of Hsp90 client proteins consistent with Hsp90 inhibition, and reduction of cellular metabolism, confirming protein degradation led to downstream effects on its signaling pathway. To round out the therapy component of our ‘see-and-treat’ paradigm, HS-27 treatment was found to reduce cell proliferation rates across breast cancer receptor subtypes.

The diagnostic component was moved into animals using a dorsal skinfold window chamber model to interrogate HS-27 uptake in vivo in the context of tumors and their surrounding microenvironment. As expected, HS-27 uptake was significantly greater in tumor window chambers than in non-tumor controls. Utilizing a fluorescent glucose analog to examine glucose uptake levels in tumors as a surrogate for aggressive disease showed that HS-27 strongly correlated (R2 = 0.96) with glucose uptake, suggesting surface Hsp90 expression is upregulated in aggressive glycolytic tumors.

To finish aim 1, HS-27 staining was performed on tumors ex vivo, achieving comparable contrast to in vivo agent administration, providing a path towards translating HS-27 to ex vivo clinical use. A small ex vivo pilot clinical study in patients undergoing diagnostic biopsy revealed a significant correlation between HS-27 uptake and the percentage of tumor present in the sample, providing first proof-of-principle of our HS-27 fluorescence-based diagnostic platform in patients. HS-27 was first imaged in biopsies in order to enroll patients across different receptor subtypes rather than at surgery where the majority of patients have estrogen receptor positive (ER+) disease. To summarize, aim 1 demonstrated a ‘see-and-treat’ paradigm in pre-clinical models of breast cancer, and provided a path towards moving HS-27 into the clinic.

With proof-of-principle patient results revealing that HS-27 may be a feasible diagnostic tool, the focus of aim 2 transitioned towards optimizing the imaging system and protocol. The ex vivo imaging strategy was optimized to minimize non-specific HS-27 uptake in preclinical models. Imaging parameters were fully vetted in a clinical study designed to interrogate HS-27 uptake in patients with breast cancer or benign conditions, as well as in a disease-free population. A high-resolution microendoscope (HRME) designed to image FITC fluorescence in a pre-clinical biopsy model was used to investigate how time between tissue excision and imaging, agent incubation time, and agent dose affect the specificity of HS-27 based diagnostics. For these experiments, a modified version of HS-27 with a 100-fold reduction in Hsp90 affinity, called HS-217, was used to establish non-specific fluorophore uptake. Calculating the ratio of HS-27 fluorescence to HS-217 fluorescence provided a ‘specificity ratio’ that was maximized with a post-excision window up to 10-minutes, 1-minute incubation time, and 100 µM dose.

The optimized protocol was then tested in 37 patients undergoing ultrasound-guided core needle biopsy and in 6 disease-free patients undergoing breast reduction mammoplasty. HS-27 uptake was significantly greater in tumor samples than mammoplasty control samples. Interestingly, HS-27 uptake was similar in tumor and benign lesion samples on average, however, examining the distribution of fluorescence across the biopsy reveals different staining patterns between tumor and benign lesions. Concurrent with the finding in aim 1 that HS-27 levels are elevated in aggressive tumors, HS-27 strongly and inversely correlated with the presence of tumor infiltrating lymphocytes, a positive prognostic marker in Her2+ and triple negative breast cancers. Additionally, leveraging both intensity and spatial patterns to generate a Gaussian support vector machine classifier allowed for accurate classification of tumor, benign lesion, and mammoplasty samples. Classification of tumor vs benign lesions resulted in an area under the receiver operating characteristic curve (AUC) of 0.93 with a sensitivity of 82% and specificity of 100%. Classification of tumor vs mammoplasty samples resulted in an AUC of 0.96 with a sensitivity of 86% and specificity of 100%.

So far, HS-27 uptake has been shown to be specific to tumor over non-tumor tissues, increased HS-27 fluorescence was suggestive of an aggressive tumor phenotype, and ex vivo HS-27 imaging accurately distinguished tumor from both benign and normal breast tissue. Two limitations of the imaging system used in aims 1 and 2 were: 1) the requirement to place the HRME probe in contact with the tissue, potentially causing artificial changes in signal due to pressure differences during probe placement, and 2) the small field of view, which prohibited translation to samples larger than 1-2 cm. Thus, the goal of aim 3 was to develop a wide-field, non-contact imaging system to demonstrate feasibility of translating ex vivo HS-27 imaging to multiple points in the breast cancer care cascade.

We have previously developed a Pocket colposcope for cervical pre-cancer detection and have recently completed construction and testing of an alpha prototype. The colposcope contains a 5 MP camera and white and green light emitting diodes (LEDs) on the tip. It weighs 1 pound, and interfaces with a phone, tablet, or computer, which provides power to the device and enables image capture. The Pocket colposcope, which will now be referred to as a Pocket mammascope is well-suited for breast margin imaging with the ability to, survey breast tumor margins as large as 10 -cm2 in a few snapshots, while maintaining the ability to image a cluster of tumor cells on a length scale of several microns.

The Pocket colposcope was modified into a Pocket mammoscope to perform fluorescence imaging through the addition of a collar with excitation LEDs and a bandpass filter for fluorescence collection. A series of bench tests show that the Pocket mammoscope can perform fluorescence imaging in a wide-field mode with a diagonal field of view of 3.25 cm (compared to 750 µm with the HRME) at a resolution of 25 µm (compared to ~4 µm with the HRME), and high-resolution mode with a diagonal field of view of 1.25 cm and resolution of 12 µm. The two imaging modes are easily navigated between through the use of a simple slider mechanism. The Pocket mammoscope was next used to image HS-27 fluorescence across in vivo and ex vivo models, with comparable results to our previous imaging systems. Additionally, the optimized ex vivo imaging protocol from aim 2 was used to shown to be compatible with the Pocket mammoscope in a cohort of patients undergoing standard-of-care ultrasound-guided core needle biopsy, and that that Pocket mammoscope is capable of imaging an entire biopsy in a single snapshot. Proof-of-concept translation to intraoperative margin assessment utilizing a window chamber model, similar to aim 1, validated that the Pocket mammoscope could image HS-27 both systemically and topically delivered to a tumor.

In conclusion, this work set out to provide a theranostic tool to reduce overtreatment for patients with breast cancer in HICs, and provide a rapid diagnostic test implementable at the point-of-care in LMICs. Towards these goals, aim 1 showed that HS-27 uptake is higher in more aggressive tumors, potentially serving as a prognostic marker delineating which patients require more or less aggressive treatment regimens. Aim 2 found that a Gaussian support vector machine classification scheme based on features from ex vivo HS-27 images accurately distinguishes tumor from both benign conditions and normal breast tissue. Finally, aim 3 demonstrated the feasibility of translating HS-27 to both diagnostic biopsy and intraoperative margin assessment by creating a Pocket mammoscope capable of imaging an entire biopsy and a tumor margin in a few snapshots. Ultimately, this work demonstrates that HS-27 imaging with the Pocket mammoscope is a means for rapid, automated detection of breast cancer, regardless of subtype, which could improve breast cancer management in both HICs and LMICs.